At Primer, we developed an ad testing method that enables you to make fast strategic decisions and optimize your budget toward proven top performers. We call it the Outlier Method.

In this post, you’ll learn what the Outlier Method is and how to use it to 3.2x your growth on paid social. We’ll discuss:

  • what to test
  • when to test
  • testing timeline
  • big ideas vs. iterations
  • how many ads to test
  • budget
  • what an outlier is

Let’s get started! (More of a visual learner? Click here to download our Outlier Method PDF!)

The Outlier Method

The Outlier Method is an ad testing methodology for paid social media. The full method lasts 2 weeks and should be repeated twice a month, if budget allows.

The goal is to optimize your ads as much as possible by testing every element: copy, creative, landing pages, and audiences. Create different versions of each to test against each other. Once you find the winners, you scale and iterate on them to help them reach their maximum potential.

This method requires a high-volume of creative testing. A/B testing is not enough to work with the Outlier Method.

Before we go into the steps of the Outlier Method, here’s an overview on how to test.

How do you test different elements on ads?

When testing an element (copy, creative, landing page, audience), keep all other elements the same for control. For example, when you test 5 new image variations, use your proven best-performing copy, landing page, and audience on all variations to see which image works best. That’s 5 tests. 

You can do the same with new copy, different audiences, new landing pages, etc..

What’s the best way to test ads?

The best way to test your ads for maximum growth is by following the Outlier Method. Here are the steps:

  1. Launch – Launch a new big idea ad set
  2. Pause – Pause ads that are underperforming
  3. Scale – Scale Outliers by adding budget and making iterations
  4. Iterate – Iterate on your winning iterations

These four ad testing steps are completed in 2 weeks and repeated twice a month. Let’s dig deeper into each step.

Step 1: Launch

Decide on a brand new idea to test. We call these “big idea tests” to differentiate from “iterative tests.”

For every big idea, you should create 3-10 visuals. Then, combine that with copy, headline, and LP combinations for a total of about 50 tests for every $25,000 in ad spend.

Day 1: Launch a new big idea ad set

Launch your tests into a separate ad set or campaign from your top-performing ads. Why? Facebook records a history on ad performance, so when you run a brand new ad against a top performer with a history of winning, Facebook will favor the top performer and your new ad test will not get a fair shot. 

On Facebook, we recommend launching ad sets with 10-25 ads in them, though you can go up to 50. On TikTok, we recommend keeping tests to between 4-6 ads per ad set.

For a partner, Farmstead, one big idea we tested was messaging around buying fresh or local in a slightly more native-style format than previous creative (at the time, native-style text was just emerging). Using Farmstead, here’s a visual example of what the creative for a big idea ad test might look like:

Creative Testing for Ads

What about budget?

For ad set budget optimization (ABO – recommended for Meta ads): set a daily budget that will yield a minimum of 3x conversions a day with your average CPA.

For campaign budget optimization (CBO – recommended for TikTok ads if monthly budget is <$100,000): pick a daily budget that ensures you can hit at least 50 conversions per week in the campaign.

For more on this, read How Many Ads You Should Be Testing Every Month to Hit Your Goals (it’s more than you think).

STEP 2. PAUSE

Day 2-4: Pause ads that are underperforming

The Outlier Method moves very fast (we don’t have time to wait for statistical significance – there’s money at stake!). Start to pause your bottom-performing ad tests immediately. 

For the first three days, watch the tests closely and quickly pause ads with higher CPAs than average. As a rule of thumb, aim to have 50% paused within 48 hours and 75% within 72 hours.

This way, you’re not wasting budget on underperformers. You’re only directing budget toward potential Outliers.

What is an Outlier?

Outliers are generally ads that perform 2-3x better than the rest in their campaign. They should have better performance than your account average by at least 10%. 

ROI of Creative Testing

STEP 3. SCALE

Day 4: Scale Outliers by adding budget and making iterations

Scale

Move those 25% that you did not pause into a campaign filled only with proven best performers. Add budget and watch them win!

Once you have 1-4 ads that either beat the previous period’s average CPA or beat your goal CPA, those are your Outliers. If you haven’t already, you’ll want to pause everything else.

Here was the best Outlier from our ad test with Farmstead:

Iterate

To get the most out of your Outliers and scale them even more, make and test at least 3 ad creative iterations for each Outlier. For example, keep the main image the same but change the copy on the image, like we did for Farmstead, below:

What is an iteration?

An “iteration” is a variation of your previous top-performing ads in which you change one or two elements, such as changing the image copy, the color scheme, or a video’s hook. By creating multiple iterations and testing those, you can thoughtfully scale your account’s Outliers.

STEP 4. ITERATE

Week 2: Iterate on your winning iterations

Keep this process going. In the second week, you’re essentially following the same process as the first week, but with iterations of Week 1’s Outliers. Yes, this means you’ll even iterate on winning iterations.

The Outlier Method should be completed within 1-2 weeks for each new big idea (including iterations). Then, depending on your budget, you should also launch new big ideas 1 – 2 times per month.

For Farmstead, the iterations mentioning “$0” and “No Fees” performed best. So, we tested these iterations next to find the highest performer of this big idea test:

That’s the Outlier Method!

Now you have all the information you need to start using the Outlier Method. Congrats! Download our Outlier Method PDF to keep this info on hand.

It’s a lot of ad testing and requires a very high-volume of creative. But since most ads fail, and because older top performers fatigue out, this high volume of ad testing is the best way to ensure long-term success and fast growth for your brand. 

You’ll need a lot of management and creative – that’s where we can help!

Your Growth Partner

If you’re looking for help scaling and testing your ads, you’ve come to the right place. We’ll handle it all for you – the creative strategy, design, production, reporting, and media buying. You just enjoy the wins.

Creative On-Demand

Need a lot of high-quality ad creative fast? Primer On-Demand is an online platform to get videos, images, and landing pages produced by top marketing designers who know what converts.